tkonolige commented on pull request #6182: URL: https://github.com/apache/incubator-tvm/pull/6182#issuecomment-669298971
Separating MKL from BLAS has two purposes: 1. We can compare other BLAS implementations vs MKL. I think MKL can be better in general, but we have no way of comparing. 2. MKL has a number of operations that are not part of BLAS like batch matrix products and sparse matrix products. With MKL separate, it is easier to add these features. ---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected]
